为 Claude Code 添加 Document Understanding

The rise of coding agents like Claude Code, Cursor, Windsurf, Cognition, Lovable, etc. marks a shift in how software is built. Instead of manually wiring APIs together, you can describe what you want through natural language, and the agent can handle the technical task of writing, executing, and iterating on the code. This opens up the possibility for “low-code IT” and allowing business users to quickly build internal and external-facing applications.

Claude Code、Cursor、Windsurf、Cognition、Lovable 等 coding agents 的崛起，标志着软件开发方式的转变。你不再需要手动把 API 拼接在一起，只需用自然语言描述需求，agent 就能完成编写、执行、迭代代码的技术任务。这为“低代码 IT”打开了大门，让业务用户能够快速构建面向内部或外部的应用。

But there’s a problem: by default, coding agents don’t natively understand documents. This limits their utility for building business applications. Enterprise applications live and breathe documents: contracts, financial reports, legal briefs, technical specifications, meeting notes. These documents are typically locked up within file formats like .pdf, .pptx, .docx, .xlsx and require specialized tooling to read and search over that information, tooling that coding agents don’t have.

但这里有个问题：默认情况下，编码代理本身并不理解文档。这限制了它们在构建业务应用时的实用性。企业应用的生命线就是文档：合同、财务报告、法律摘要、技术规范、会议记录。这些文档通常被锁在 .pdf、.pptx、.docx、.xlsx 等文件格式里，需要专门的工具才能读取并检索其中的信息，而编码代理并不具备这些工具。

This may sound surprising at first glance. But coding agents have real limitations for understanding files:

乍一听可能令人惊讶，但编码代理在理解文件方面确实存在局限：

Cursor doesn’t support PDF upload at all (and many other files).
Cursor 完全不支持 PDF 上传（也不支持许多其他文件）。
Claude Code has a Read capability that has basic PDF understanding capabilities, but a max file size of 32 MB and 100 pages per request.
Claude Code 具有 Read 能力，具备基本的 PDF 理解功能，但每个请求的文件大小上限为 32 MB，页数上限为 100 页。

By equipping coding agents with the right tools around document understanding:

通过为编码代理配备围绕文档理解的正确工具：

They can pull in more context. This means building apps that better adapt to business requirements.
它们可以拉取更多上下文。这意味着构建的应用能更好地适应业务需求。
They can use the tools within the generated code. This means building apps that are more agen...